Apdullah Yayik

NEURAL NETWORK BASED CRYPTOGRAPHY

In this paper, neural network based cryptology is performed. The sys- tem consists of two stages. In the first stage, neural network-based pseudo-random numbers (NPRNGs) are generated and the results are tested for randomness us- ing National Institute of Standard Technology (NIST) randomness tests. In the second stage, a neural network-based cryptosystem is designed using NPRNGs. In this cryptosystem, data, which is encrypted by non-linear techniques, is subject to decryption attempts by means of two identical artificial neural networks (ANNs). With the first neural network, non-linear encryption is modeled using relation- building functionality. The encrypted data is decrypted with the second neural network using decision-making functionality.

Orthogonal Extreme Learning Machine Based P300 Visual Event-Related BCI

Brain Computer Interface BCI is a type of human-computer relationship research that directly translates electrical activity of brain into commands that can rule equipment and create novel communication channel for muscular disabled patients. In this study, in order to overcome shortcoming of Singular Value Decomposition in Extreme Learning Machine, iteratively optimized neuron numbered QR Decomposition technique with different approaches are proposed. QR Decomposition Extreme Learning Machine technique based P300 event-related potential BCI application that achieves almost % 100 classification accuracy with milliseconds is presented. QR decomposition based ELM and novel feature extraction method named Multi Order Difference Plot MoDP techniques are milestones of proposed BCI system.

Improving Pseudo random number generator using artificial neural networks

Pseudo-random number generators generate sequent of digits that cannot be expected before. Random number generators are used in lots of studies especially physical and statical implementations. In this paper; by using Multi-Layer Perceptron Neural Network, a traditional random number generator is strengthened. In the end of the study; both of random number generators are tested by some randomness tests of National Institute of Standard Technology test suite. As a result, it is learned that Neural Networks can generate good random numbers.

Diagnosis of congestive heart failure using poincare map plot

In this study, in order to diagnose congestive heart failure patients (CHF), Poincare map obtained from raw ECG data is used. CHF and normal ECG data, which are distributed freely via internet, are analyzed. Poincare map is divided into equal rectangle cells and points in all of the cells are determined. These values are used for knn (k-nearest neighbour) classification. At the result of this study, it is considered that CHF patients and normal people can be separated each other using features obtained from Poincare map of Raw ECG record.

Brain computer interface based visual detection system

In this study, pattern recognition based brain computer interface is designed using EEG p300 component elicited by visual stimuli. A novel EEG database obtained from 19 subjects is constructed with EMOTIV EPOC+ amplifier and OPENVIBE software. Extreme Learning Machine, a type of single layer neural network, Λ-nearest neighbour, Bayesian network and Multi-Layer Perceptron classifiers are compared for classification task in terms of training duration and performance measurements. As a result of subject-based classification, it is observed that Extreme Learning Machine classifier is more efficient and useful.

Patient Specific Congestive Heart Failure Detection From Raw ECG signal

In this study; in order to diagnose congestive heart failure (CHF) patients, non-linear second-order difference plot (SODP) obtained from raw 256 Hz sampled frequency and windowed record with different time of ECG records are used. All of the data rows are labelled with their belongings to classify much more realistically. SODPs are divided into different radius of quadrant regions and numbers of the points fall in the quadrants are computed in order to extract feature vectors. Fishers linear discriminant, Naive Bayes, Radial basis function, and artificial neural network are used as classifier. The results are considered in two step validation methods as general k-fold cross-validation and patient based cross-validation. As a result, it is shown that using neural network classifier with features obtained from SODP, the constructed system could distinguish normal and CHF patients with 100% accuracy rate.